Code for paper: Attribute-Aware Attention Model for Fine-grained Representation Learning
Clone or download
Latest commit ab32107 Jan 3, 2019
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
fig update readme Nov 17, 2018
model init commit Oct 5, 2018
tools update readme Nov 17, 2018
.gitignore add attribute file Nov 17, 2018
CUB.py add attribute file Nov 17, 2018
cub_demo.py add attribute file Nov 17, 2018
readme.md update readme Jan 3, 2019
run.sh add attribute file Nov 17, 2018

readme.md

Attribute-Aware Attention Model

Code for ACM Multimedia 2018 oral paper: Attribute-Aware Attention Model for Fine-grained Representation Learning

We have presented fine-grained classification, person re-id, image retrieval in the paper. Here is the example of fine-grained classification. For detailed results, refer to the original paper or ArXiv.

Usage

Requires: Keras 1.2.1 ("image_data_format": "channels_first")

Run in two steps:

  1. Download CUB-200-2011 dataset here and unzip it to $CUB; Copy file tools/processed_attributes.txt to $CUB.
  • The $CUB dir should be like this:

  1. Change data_dir in run.sh to $CUB, run the scprit sh run.sh to obtain the result.
  • Result on CUB dataset

Citation

Please use the following bibtex to cite our work:

@inproceedings{han2018attribute,
  title={Attribute-Aware Attention Model for Fine-grained Representation Learning},
  author={Han, Kai and Guo, Jianyuan and Zhang, Chao and Zhu, Mingjian},
  booktitle={Proceedings of the 26th ACM international conference on Multimedia},
  pages={2040--2048},
  year={2018},
  organization={ACM}
}